Week 1, 01/11/17: What is (are) data? epistemology & research design

Topics:

Data, information, knowledge production — different definitions & epistemic positions

Readings for this Class:

Notes on Readings:

**Floridi, Information a Very Short Introduction — Chapter 2:**

Data is a difficult concept to define. However, it is widely considered something that can be manipulated (think ‘data mining’ and ‘information management’). According to the General Definition of Information (GDI) there are three parts to that define data. GDI.1 states that information is made of data. GDI.2 states that the data are well-formed, meaning they follow the appropriate syntax for the chosen system. GDI.3 states that the well-formed data are meaningful. According to Floridi, the important takeaway of GDI.3 is that the data that forms the information can be meaningful, regardless of the audience. In other words, I might not be able to understand the meaning of the information, but someone does and therefore it can be considered meaningful.

Data doesn’t require a format like words or symbols to convey information. Sometimes the absence of something conveys the information. Like silence coming from a child’s room. A parent usually assumes that something is amiss. This concept is summarized in Floridi’s statement that “Datum is ultimately reducible to a lack of uniformity.” Or, as Donald MacCrimmon MacKay noted, “Information is the distinction that makes a difference.”

Drucker, “Humanities Approaches to Graphical Display.”

Drucker makes an argument for defining data that contrasts Floridi’s definition. Drucker argues that data is constructed rather than given and is entirely dependent on the point of view of the observer. According to Drucker, humanists need to break away from the overly-simplified constructs of standard statistical collection, analysis and visualization of data and go deeper to emphasize complexity of the various layers.

I agree with Drucker that standard visualizations of statistical data are often simplified versions of more complex data. Every depiction of data has some form of bias. I don’t think that Drucker’s argument for interpreting data through a humanist lens lessens that bias in any form. It just introduces a different set of biases to determine which data points to display. It does not address the fact that analysts and information designers need to make arbitrary decisions along the way in order to give form to data for an audience to interpret.

Class Exercises:

What is (are) fake news? Googledoc by Kate & Maeve

Notes from Class:

The way we interpret or understand data can have consequences.

What is Data?

uninterpreted information
“raw material”
recorded information
observation
unstructured (or better unorganized?) / unprocessed
characteristics
datum = reducible to a lack of uniformity
“well-formed,” formatted

Data vs. Information

information is made of data
information = meaningful data

Facts

“evidence” — implies a claim of truth

Knowledge

embodiment — implies someone who knows
some level of abstraction and systematization

What makes a good research question?

Leads to new knowledge (significance)
is precise
Can be answered (but is not trivial)
its answer leaves room for analysis and explanation
Is connected to bigger issues (can be generalized)
avoid yes / no questions (unless you are testing a given hypothesis)
broad vs. narrow questions
skinny vs. fat questions

Terms:

Methodology: The study of methods.
Ontology: The branch of physics dealing with the nature of being.
Epistemology: The theory of knowledge. The investigation of what distinguishes justified belief from opinion.
Positivism/Constructivism: Both are viewed as epistemologies that present a different idea of what constitutes as knowledge. Positivism can be understood as a philosophical stance that emphasizes that knowledge should be gained through observable and measurable facts (Realists). In this sense, this is considered as a rigid scientific inquiry. On the other hand, Constructivism states that reality is socially constructed (Anti-Realist). That everything is a simulated perception.